AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Vulnerability-Aware Robust Multimodal Adversarial Training

Junrui Zhang, Xinyu Zhao, Jie Peng, Chenjie Wang, Jianmin Ji, Tianlong Chen

AAAI 2026 · Adversarial robustness and security › Certified and empirical robustness · approach: Robustness-accuracy trade-off analysis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add VARMAT gradient-norm regularizationFGSM-RS · CMU-MOSEI PGD lambda=0.5robust accuracy: 55.76 -> 68.49Improved
add VARMATFGSM-PCO · UR-FUNNY PGD lambda=0.5robust accuracy: 41.21 -> 59.55Improved
add VARMATFGSM-EP · AVMNIST PGD lambda=0.5robust accuracy: 0.35 -> 11.08Improved
add VARMATFGSM-EP · CMU-MOSEIclean accuracy: 79.09 -> 78.07Hurt
add VARMATFGSM-RS · UR-FUNNYclean accuracy: 65.69 -> 63.80Hurt
add VARMATFGSM-RS · AVMNISTclean accuracy: 69.38 -> 63.21Hurt
add VARMATFGSM-RS · CMU-MOSEIclean accuracy: 78.83 -> 79.15Improved
regularize all modalities instead of only the most vulnerable (vs S1) (ablation)VARMAT FGSM-RS · CMU-MOSEI PGD lambda=0.5robust accuracy: 62.63 -> 68.49Improved
regularize instead of reallocating perturbation budget in attack (vs S2) (ablation)FGSM-RS · CMU-MOSEI PGD lambda=0.5robust accuracy: 35.54 -> 68.49Improved
regularize all modalities instead of attacking only the most vulnerable (vs S3) (ablation)FGSM-RS · CMU-MOSEI PGD lambda=0.5robust accuracy: 51.88 -> 68.49Improved
exclude feature norm from regularizer (vs Trap) (ablation)FGSM-PCO · input-space PGD lambda=0.1robust accuracy: 2.76 -> 53.67Improved
exclude feature norm from regularizer (vs Trap) (ablation)FGSM-RS · CMU-MOSEI feature PGD lambda=0.5robust accuracy: 75.83 -> 68.49Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy